【发布时间】：2020-08-20 16:51:20
【问题描述】：
尽管之前有 R 和 Python 的编码经验，但我对 Jupyter Notebook、tensorflow 和深度学习模型构建是全新的，所以我正在寻找帮助我诊断此错误的人。我正在学习一个演示如何使用深度学习模块对图像进行分类的教程 (https://www.youtube.com/watch?v=wQ8BIBpya2k&list=PLQVvvaa0QuDfhTox0AjmQ6tvTgMBZBEXN)。模型加载 mnist 图像数据集，并将图像分类在 1-10 之间。运行三个 epoch，整体模型准确率达到 97%。
#Import module
import tensorflow as tf
#Import image dataset
mnist = tf.keras.datasets.mnist
(x_train, y_train), (x_test, y_test) = mnist.load_data()
#Defining training and test
x_train = tf.keras.utils.normalize(x_train, axis = 1)
x_test = tf.keras.utils.normalize(x_train, axis = 1)
#Define model
model = tf.keras.models.Sequential() #Feed-forward model
#Define imput layer
model.add(tf.keras.layers.Flatten()) #Input layer
#Two hidden layers
model.add(tf.keras.layers.Dense(128, activation=tf.nn.relu)) #Rectify linear, default
model.add(tf.keras.layers.Dense(128, activation=tf.nn.relu))
#Output layer
#Corresponds to the number of classifications; ten in this case
#No relu because it is a probability distribution
model.add(tf.keras.layers.Dense(10, activation=tf.nn.softmax))
#Defining training parameters for the model
#Loss is error, or what you have gotten wrong
model.compile(optimizer = 'adam',
loss= 'sparse_categorical_crossentropy', #could use binary if looking for cats/dogs
metrics = ['accuracy'])
#Training the model
#Epoch = how many times the model runs
model.fit(x_train, y_train, epochs = 3)
输出如下：
Epoch 1/3
1875/1875 [==============================] - 1s 578us/step - loss: 0.2612 - accuracy: 0.9236
Epoch 2/3
1875/1875 [==============================] - 1s 571us/step - loss: 0.1068 - accuracy: 0.9668
Epoch 3/3
1875/1875 [==============================] - 1s 562us/step - loss: 0.0721 - accuracy: 0.9773
当我试图找出模型损失时......
#Printing the model loss
val_loss, val_acc = model.evaluate(x_test, y_test)
print(val_loss, val_acc)
...我收到此错误：
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-6-3452f7a38776> in <module>
----> 1 val_loss, val_acc = model.evaluate(x_test, y_test)
2 print(val_loss, val_acc)
~\AppData\Roaming\Python\Python36\site-packages\tensorflow\python\keras\engine\training.py in _method_wrapper(self, *args, **kwargs)
106 def _method_wrapper(self, *args, **kwargs):
107 if not self._in_multi_worker_mode(): # pylint: disable=protected-access
--> 108 return method(self, *args, **kwargs)
109
110 # Running inside `run_distribute_coordinator` already.
~\AppData\Roaming\Python\Python36\site-packages\tensorflow\python\keras\engine\training.py in evaluate(self, x, y, batch_size, verbose, sample_weight, steps, callbacks, max_queue_size, workers, use_multiprocessing, return_dict)
1354 use_multiprocessing=use_multiprocessing,
1355 model=self,
-> 1356 steps_per_execution=self._steps_per_execution)
1357
1358 # Container that configures and calls `tf.keras.Callback`s.
~\AppData\Roaming\Python\Python36\site-packages\tensorflow\python\keras\engine\data_adapter.py in __init__(self, x, y, sample_weight, batch_size, steps_per_epoch, initial_epoch, epochs, shuffle, class_weight, max_queue_size, workers, use_multiprocessing, model, steps_per_execution)
1115 use_multiprocessing=use_multiprocessing,
1116 distribution_strategy=ds_context.get_strategy(),
-> 1117 model=model)
1118
1119 strategy = ds_context.get_strategy()
~\AppData\Roaming\Python\Python36\site-packages\tensorflow\python\keras\engine\data_adapter.py in __init__(self, x, y, sample_weights, sample_weight_modes, batch_size, epochs, steps, shuffle, **kwargs)
280 label, ", ".join(str(i.shape[0]) for i in nest.flatten(data)))
281 msg += "Please provide data which shares the same first dimension."
--> 282 raise ValueError(msg)
283 num_samples = num_samples.pop()
284
ValueError: Data cardinality is ambiguous:
x sizes: 60000
y sizes: 10000
Please provide data which shares the same first dimension.
我可以做些什么来修复这个错误并计算损失？
【问题讨论】：
标签： python tensorflow keras deep-learning jupyter-notebook